Deep neural network-based baby cry identification method and system

Abstract

The invention provides a deep neural network-based baby cry identification method and a deep neural network-based baby cry identification system. The method comprises the following steps of acquiring baby cry data for training; performing classification and labeling on the baby cry data for training; extracting a Mel-domain cepstrum coefficient of each segment of audio in each piece of classified and labeled baby cry data for training to generate a data file for training; obtaining an initial weight of each layer in a deep neural network in a layer-wise pre-training way according to the data file for training; acquiring a deep neural network-based cry model according to the initial weights of all the layers in the deep neural network by virtue of a BP (back-propagation) algorithm; acquiring baby cry data to be identified, and extracting a Mel-domain cepstrum coefficient of each segment of audio in the baby cry data to be identified; performing cry identification according to the Mel-domain cepstrum coefficient of each segment of audio in the baby cry data to be identified and the cry model. According to the method and the system, the baby cry identification rate can be increased.

Description

technical field [0001] The invention relates to a method and system for recognizing baby crying sounds based on a deep neural network. Background technique [0002] The research on baby crying recognition has appeared since the 1960s and 1970s. Limited by the previous technical level and data scale, there are not many products and technologies with application prospects, and most of the products on the market have unreliable recognition performance and technical problems. Low content and other shortcomings. On the one hand, this is due to the small scale of baby crying data collection and labeling in history, some even have only dozens of audio segments, and usually only effective identification of crying types with large differences, such as healthy babies and deaf children The recognition of crying sounds is difficult to fully explore the laws behind the crying of babies, and the reliability of distinguishing more states is not high; However, the modeling ability is limi...

AI Technical Summary

Problems solved by technology

[0002] The research on baby crying recognition has appeared since the 1960s and 1970s. Limited by the previous technical level and data scale, there are not many products and technologies with application prospects, and most of the products on the market have unreliable recognition performance and technical problems. Disadvantages such as low content

On the one hand, this is due to the small scale of baby crying data collection and labeling in history, some even have only dozens of audio segments, and usually only effective identification of crying types with large differences, such as healthy babies and deaf children The recognition of crying sounds is difficult to fully explore the laws behind the crying of babies, and the reliability of distinguishing more states is not high; The modeling ability is limited, and the baby's crying sound cannot be fully modeled, so the recognition rate of the finite state is not high, and there are few successful applications.

Images